In this study, among outpatients and outpatients with UC, we compared BWT by TUS using CS findings in each segment of the colon, and showed that BWT reflected mucosal inflammation. In addition, a positive BWT of the colon accurately determined the presence of mucosal inflammation in UC. Therefore, TUS can be a useful and suitable method for detecting mucosal inflammation in UC for both inpatients and outpatients.
TUS has been a useful tool for evaluating the location of mucosal inflammation and assessing the disease activity of UC.9,12,17,21,22 Among the various studies on UC, some have focused on assessing disease extent and location with a detailed consideration of mucosal inflammation.9,12,17,21 These studies had a positive BWT of more than 3 mm or more, which may include areas with mild mucosal inflammation. Additionally, MES 1 was not classified as mucosal inflammation. There is a recent trend regarding MES 0 as mucosal healing, and there is a need for a stricter assessment of mucosal states.14,15 The improvements of US machines have improved their spatial resolutions in comparison to past models, allowing the possibility of a more detailed observation of intestinal states. Therefore, we believe conventional intestinal evaluation with TUS was not sufficient to evaluate mucosal inflammation and consequently changed the evaluation method and analyzed it.
Our results indicated that BWT of UC patients tended to increase with increasing endoscopic severity, suggesting that measurement of BWT by TUS would be useful for the evaluation of mucosal inflammation. Regarding the correlation between BMT and MES, although there was a difference depending on the site, a high correlation was detected on the right colon side. Since it is difficult for patients with severe conditions to undergo a total CS examination, these results suggest that the degree of inflammation on the oral side might be evaluated noninvasively to some extent. On the other hand, in another report, the left colon was found to have a higher correlation for BMT and MES.20 This may be due to differences in patient background, such as the intensity of inflammation and the number of patients analyzed as well as in the variations in the definition of BMT of the positive findings. It is recommended that further studies are necessary, including studies at other facilities, to increase the sample size and to clarify further the correlation between BMT and MES.
Regarding BMT, the optimal BWT varied slightly from segment to segment. However, our data suggest that BWT >2 mm as a positive finding might be appropriate for detecting mucosal inflammation. With this criterion, the calculated specificity of each site tended to have a lower value, but the results were almost comparable to those reported previously.12,23,24 When the positive BWT was >3 mm, the specificity increased; however, the sensitivity decreased. As a result, BWT >2 mm as positive findings performed better overall. In our study, cases of mucosal inflammation of 2 mm or less and the opposite cases were only occasionally encountered. The reason may be that there are individual differences in wall thickening, and that the evaluation was performed in a very narrow range of mm units, which may be prone to measurement errors by the performing physician. Therefore, evaluation by multiple physicians is necessary.
We also examined the presence of MES >1 as the basis of mucosal inflammation of each segment. Based on the results, the sensitivity was acceptable, but the specificity was extremely low. As such, there may be an increase in the possibility of overlooking the presence of inflammation. The data also suggested that our evaluation method could clearly detect mucosal inflammation. Since the evaluation of mucosal blood flow measurements in patients with UC using the color Doppler method has been reported,12,20,25,26 evaluating this method, in combination with BWT, may also improve accuracy and stratify the degree of mucosal inflammation, so it may be an examination to be considered in the future.
This study has some limitations. One is that the number of cases at each site varies, and the oral intestinal tracts, such as the ascending and transverse colon, have numerous normal mucosa, and the number of inflammatory mucosa increases toward the anal sides. One of the causes of this imbalance could be that we mixed patients with different phenotypes (for example, 82% of patients had pancolitis and 96% of patients enrolled had an active state). This could have led to overestimation of the diagnostic accuracy. Moreover, this imbalance was also thought to be the cause of the dissociation between the accuracy rate and AUC, and the difference in correlation between BWT and endoscopic activity in each segment. To address this, the study targeted a wider patient population (from outpatients to inpatients with clinically severe cases, including those who are hospitalized or those who are in remission) to make the study more in line with real general practice. UC is characterized by its continuity, extending from the rectum to the mouth1, and it is difficult to increase the number of evaluations of inflammatory mucosa, especially in severe states of the oral colon because in some cases, there is a risk of performing endoscopy itself. With regard to the distal side, it may be possible to solve this problem by increasing the number of examinations in patients in clinical remission states. Furthermore, blinding of tests is also necessary. On the other hand, considering the fact that patients with clinically severe states were also included, AUS images may be useful for assessing severity if other characteristics, such as blood flow, are considered, in addition to wall thickness. However, the range was limited to the observable distal colon on endoscopy.
Another limitation is that this study involved a retrospective analysis. Therefore, there were some various biases, which may be attributed to the possibility of additional treatment modification, incorrect examination period, non-blinding of tests, and the cases not being evaluated by multiple individuals.
Regarding the change in BWT related to the additional treatments and incorrect examination period, a recent report showed that BWT improved in two weeks after successful treatment.27 Therefore, in cases where drugs with relatively immediate effects, such as infliximab or tacrolimus, were used, the effects of the drug might affect TUS findings if the examination period was prolonged. On the other hand because there was no contrast with endoscopic findings, it was unclear how much the two examinations would affect the results in about two weeks. In our cases, a few patients had an examination period of 14 days, but the clinical findings, such as stool frequency and blood data, did not change significantly, so they were included in this study. In addition, there were no cases of endoscopic remission or TUS-negative findings due to the effects of treatment during the examination period, and no inflammation findings remained. Therefore, we believe that there would be no significant effect on the examination of the presence or absence of inflammation.